Cytomegalovirus (CMV) infections remain a serious problem in hematopoietic stem cell transplant patients. Following transplantation CMV infections can cause peumonititis, hepatitis, enteritis, and marrow failure. CMV seropositive transplant recipients can be treated with antiviral agents such as ganciclovir at the onset of infection or at the time of stem cell engraftment, but ganciclovir therapy is associated with renal toxicity and suppression of neutrophil counts. Preliminary studies have found that adoptive immune therapy using CMV-reactive cytotoxic T lymphocytes (CTL) is an effective and less toxic alternative to prevention of CMV infection in transplant recipeints. The purpose of this study is to develop new treatment strategies for producing CMV-reactive CTLs that can be used for adoptive immunetherapy and to better understand the cellular immune response to CMV. Current studies are focused on identifying the immune dominant peptides that can be used to stimulate CMV reactive CTLs. CMV contains over 200 proteins, but two proteins CMV proteins pp65 and IE1, have been found to be immune dominant. We found that the peptide CMV pp65 91-100 was found an immune dominant peptide restricted to HLA-A*33 and pp65 328-337 was an immune dominant peptide restricted to HLA-A*2402. We are now using libraries of pp65 and IE1 overlapping peptides of 15 amino acids in length to identify new class I and class II epitopes. Class I and class II epitopes have been found among both the pp65 and IE1 peptides. To further characterize the class I eptitopes, the 9 amino acid peptides that made up the reactive 15 amino acid peptides were tested for reactivity toward CD8 cells. Several pp65 and IE1 reactive peptides 9 amino acids in length have been identified. Future studies will characterize the HLA restrictions of the class I and class II epitopes.